Security fixes are applied to the latest release and the default branch.

Use GitHub private vulnerability reporting for prompt-injection bypasses, unsafe data disclosure, credential exposure, dependency compromise, workflow vulnerabilities, or other security-sensitive issues.

Do not open a public issue or attach a real sensitive document. Include:

• affected version and environment;
• a minimal synthetic reproduction;
• expected and observed behavior;
• impact and realistic attack conditions;
• any proposed mitigation.

• Store API keys in .env, a secret manager, or environment variables.
• Never embed a key in code, examples, command output, issues, or test fixtures.
• Use a restricted OpenAI project key and review usage limits.
• Rotate a key immediately if exposure is suspected.

Model output is probabilistic and may be incorrect. Prompt-injection defenses reduce risk but cannot guarantee that arbitrary hostile documents are safe. Applications making consequential decisions must add domain-specific validation, authorization, human review, and output handling.